Each year in the United States alone, hundreds of millions of red blood cell (RBC) antigen typings are performed on donated units of blood and the patients that are to receive them. In addition, equivalent numbers of patient antisera are screened for the presence of pre-existing anti-RBC antibodies, the specificities of which must be identified prior to the selection of compatible blood. The technology used in blood banks for doing these tests is essentially the same as the one demonstrated by Landsteiner over 100 years ago—the agglutination of RBCs by an appropriate antisera. Assay systems of this type are labor intensive and typically require teams of highly-trained medical technologists manually shaking test tubes over magnifying mirrors and assessing agglutination patterns by eye. Consequently, blood banks require significantly more bench technologists per test than any other type of clinical laboratory, as reflected in the 10- to 100-fold greater cost per test for the transfusion laboratory than those for other areas of laboratory medicine. In addition, blood donation facilities, blood banks, and hospital transfusion services across the country are facing a growing shortage of skilled staff to perform such tests due to the lack of qualified and interested candidates. This is particularly concerning given the extraordinary importance of accurate pre-transfusion testing and the ability to provide blood components to patients in a timely, often emergent, basis.
As opposed to other forms of laboratory testing such as those in clinical chemistry, coagulation, and hematology, blood bank testing has defied the development of rapid, high-throughput automation. The methods for blood bank automation that are currently available require, in essence, the use of a machine that detects the agglutination of red cells, but agglutination (or some variant thereof) is still the end-point much as it was nearly 100 years ago. Reasons for the difficulty in developing truly automated blood typing systems are multiple, but in large part have to do with the need to work with intact cells in order to detect the presence of specific polymorphic molecules on their surfaces. This is in contrast to other laboratory tests that simply count numbers of cells or measure the concentrations of soluble plasma proteins or electrolytes.
While it is true that flow cytometric testing also detects cell-surface phenotype, the indications for such tests do not, in general, require rapid real-time results such as those required in transfusion medicine where the goal is to prevent the transfusion of incompatible blood, often during emergencies such as trauma or unanticipated surgery, where time and accuracy are of the essence. Furthermore, essential differences in the nature of blood bank testing have precluded the development of “point-of-care” testing devices, such as those now available for glucose or electrolyte determinations or for the rapid “on-the-scene” diagnosis of myocardial infarction. The development of novel blood bank testing methods could lead to the development of small, portable devices for pre-transfusion testing that could facilitate “point-of-care” (e.g., battlefield) testing not possible using conventional approaches.
Another significant issue in blood banking testing is the growing unavailability of complete panels of high-quality immunological reagents for typing. Supplies of conventional sources come from donated human polyclonal antisera that are difficult to quality control and are dwindling in supply due to growing ethical concerns regarding the deliberate hyperimmunization of reagent donors. Because immune responses to many blood group antigens are mounted only in humans (who lack the particular antigen) and not in animals (e.g., mice, whose immune systems generally cannot detect the subtle human polymorphisms to which the antisera needs to be directed), efforts to produce monoclonal typing reagents have required the ability to transform human B-cells, which is a very inefficient and expensive endeavor. Therefore, the availability of endless supplies of well-characterized monoclonal RBC antibodies, analogous to those which revolutionized the automation of other immunological-based assays, such as those for endocrinology or infectious diseases, has been problematic in the field of transfusion medicine.
More than 20 million units of blood are collected in the United States annually, with worldwide collections exceeding 40 million units. Blood collection centers (e.g., American Red Cross, hospital-based donor centers), hospitals, and other blood banks and transfusion centers all have on-going needs to type blood quickly and accurately in a high-throughput manner. Small, automated, blood typing instruments would also have “point-of-care” applications in physician offices such as those of obstetricians in which a patient's Rh type needs to be determined in order to properly administer Rh(D)-immune globulin. Each unit of blood that is collected is typed for at least 3 (i.e., A, B, Rh(D)) antigens and often the blood is tested for detection of many more antigens (e.g., Rh(C), Rh(c), Rh(E), Rh(e), K, Fya, Fyb, M, N, S, s, Jka, Jkb, and the like).
Upon receipt of units by a blood bank, standards require that each unit be retested for A and B to ensure proper labeling. Each collected unit of blood is separated into red cells, platelets, and plasma in order to treat 3 different patients with different needs. Approximately twice as many patients are typed for A, B, and Rh(D) (and often other antigens) than those who actually receive blood (i.e., crossmatch/transfusion ratio is approximately 2). In addition, blood samples are collected every seventy-two hours on hospitalized patients in order to have fresh samples available for cross-matching purposes such that many patients are typed and retyped many times during their hospitalization. Therefore, the number of blood typings performed worldwide annually is in the hundreds of millions of tests.
As noted previously, essentially all methods for RBC typing, whether manual or automated, use agglutination as the endpoint. The disadvantages of manual methods include labor costs, low throughput, and human error. Disadvantages of current automated methods include inability to multiplex testing reactions and relatively low throughput when compared to other laboratory testing. Additionally, significant disadvantages of both current manual and automated methods include their reliance on conventional sources of antisera, which sources are dwindling in supply and can potentially transmit human disease, or the few human or mouse hybridoma-produced antibodies which are difficult and expensive to produce. The present invention provides endless supplies of inexpensive phage-displayed anti-RBC reagents that can be used not only in an automated “phenotyping-by-reagent genotyping” technology as disclosed herein, but that are also compatible with conventional manual and automated agglutination methods using anti-M13 antibody as the agglutinating (i.e. “Coombs”) agent (e.g., U.S. Pat. No. 5,985,543, to Siegel).
In sum, there is a long-felt and acute need for improved blood typing methods and reagents therefore, which will allow the automation of such tests thereby lowering costs, improving efficiency and accuracy, and obviating the need for current difficult to obtain reagents. The present invention meets these needs.